Do these consoles sound different?
ฝัง
- เผยแพร่เมื่อ 23 มิ.ย. 2023
- We tested 13 popular audio consoles and two I/O devices to see if they sound different, and by how much. The results were surprising!
We want to give a special thank you to John Harrel, Robert Shriver, and Frank Hernandez for letting us borrow gear for this test. - วิทยาศาสตร์และเทคโนโลยี
If you can team up with @DaveRat for the next one, I think the two of you will blow every preconception everyone has about consoles.
Dave, you there?
Dave Dave, you there?
@@devinlsheets_alphasound Dave's not here, man
@@jsalvatori That's a deep cut.
@@aaronaustin7760 it didn't used to be, but I'm old.
Great deep dive! Thanks for taking the time and effort to test and share this!
It's nice to hear someone, who knows infinitely more about the tools we use, shed light like this. What a massive effort! Fantastic job!
Fantastic video. A laser-close comparison with interesting tests, thank you!! Well done, hope you're able to do many more of these!
My friend, those 3 Yamaha consoles that when setting their compressors to "0" attack, they created distortion. Well, I applaud these units, as this is EXACTLY how they SHOULD behave when approaching attacks of 0.0000x seconds. My working knowledge of 26+yrs as a contract engineer specializing in multiband audio processing for the broadcast industry has afforded me the experience to say that Yamaha has done something truly unique with their onboard DSPs. When setting a compressor to that fast of a clamping action (attack), you've essentially created something more than a compressor. Next in line (amplitude control-wise) would be a limiter. After that, having the greatest amount of amplitude control would be a 'Clipper', which grabs the audio signal so quickly that no dynamic peaks can occur beyond its threshold point. The very well known downside to this accuracy of control is distortion. And TONS of it. Through this process harmonic noise is created in multitudes, odd/even and everything in between.
Notice how the overall initial impulse was reduced by what would appear to be maybe -¾ of the original impulse amplitude? Did anyone notice this??
This in itself shows that these three devices were the ONLY ONEs capable of doing the exact job which they are supposed to do! and, they did this with a stunning ability. This Unfortunately, does result in what is known as "distortion". Due to the fact that the amount of reduction of the impulse signal which was introduced, I would dare say these were fast enough to be considered a literal clipping type limiter. The fact that these resulted in distortion is exactly what they should have done. Especially when you consider the speed that they are released. A fast release, something along the line of less than 100 milliseconds will certainly result in a distorted signal. However, try this again with a release somewhere on the line of 250 maybe 500 milliseconds, or even one full second and the distortion will probably have been completely smoothed out considerably. But in my book, as well as my line of work should dictate, these were the only three devices able to make a reduction in overall amplitude upon that impulse signal. All the rest of the ones were just complete rubbish useless DSPs just meant to smoothly squish a sound. This is not accurate clipping or compression. However again, I applaud Yamaha. Absolutely amazing work for a onboard DSP!!!!
Woah, I've never seen such an elaborate test! Thanks for sharing your knowledge and perspective. I'd like to agree with you about the Digico compressor, it sounds pretty rough. Can't wait for part 2!
I ve been waiting for years for a review like this . Thank you !
Brilliant video and setup. Well done, and thank you.
Killer job on this, man! I'm getting my hands on a DM7 this week and hope to do something similar and put it through its paces. Will be interesting to see if the processing is the same or similar to the DM3.
Incredibly informative. Thanks for sharing this!
I am really impressed about the work you put in for this video!!! Thanks!
I appreciated the way you noted console eq/dynamics can react so differently given "the same" adjustments. Some consoles I mix on - even very expensive ones - have driven me a bit crazy but I thought because they are expensive, it must be my problem or lack of understanding.
It's always been that way. Just use the EQ on an old soundcraft FX16 and then jump to an A&H PA12. The EQs on these old inexpensive analog consoles feel quite different. It might be well on purpose that in the digital domain, manufacturers try to retain their style of EQ from the analog era.
This is fantastic stuff, more like this please
Astonishing! Thanks for the great effort!
Omg this is a brilliant work! Huge thanks!
Great video! Would be nice to see how preamps react to relatively high impedance microphones in the part two.
so ProSoundTraining did a Digital Mixer Study, you can look at the Timing tests and see that the A/D converters in the x32, M32, and midas Pro2/DL251 all show mostly post-ringing of the impulse test, so i'm fairly certain the A/D converters are running "Minimum-phase" mode which has lower latency, but increases phase differences at high frequencies. Honestly I prefer the phase coloration it adds.
Haven’t even watched the full video yet - but I’d love to see something from the SSL lineup added in here!
good video, your efforts are well appreciated!
Thank you so much! This is why I’m on youtube. Learned a lot! ❤❤
Wow. Thank you for all that hard work.
Interesting and complex way to test the differences between microphones... ;) I would have liked to see the preamp null comparisons done by aligning the IR of each console into your dual channel FFT analyzer and then compare the results of the actual preamps without introducing the imperfect transducers into the equation. Interesting results with the compressor tests, though. I think that a follow up could be done with measuring preamp noise, distortion when pushed hard, and what the results of summing a sine wave are; tricky, I know. Just curious to see if this test would reveal anything useful...
Lastly, I'd LOVE to see a Midas Pro series desk. Their channels always sound compressed even when the compressor is not showing any GR.
Amazing Work thank you!
Great video. Regarding nulling the signal maybe it might have made a difference if you had a separate unit for clocking and sent the clock to all consoles simultaneously ?
Aaesome! Have you did samples recording through that consoles? Interesting to examine uncompressed files using statistic methods. If yes, could you please upload uncompressed wav’s?
Please do a part 2 and possibly, include Yamaha DM7, Soundcraft, Behringer Wing, Presonus StudioLive. Great video & Thanks in advance!
This Is awesome! my only comment Id make with the compression section is that for the Allen and heath compressor you used the RMS compressor not the PEAK compressor. On the other consoles you used the PEAK compressors. For the test that you performed you'd get better results using the PEAK over RMS.
As you mentioned the compressors seem to be the most different in each console. I wonder how the "premium effects" or DEEP plugins sound compared to the stock EQ and Compression in each desk
yeah also waves/avid stuff etc
absolutely, the Digico blows everything out of the water imo. It's the only compressor that brings out the ambience in the sample in a pleasing way.
THIS
Re the first part of your test: you'll want to get the Eventide Precision Time Align for this kind of stuff...it allows you to do sub-sample time delay...I wouldn't be surprised if you could get them all to null to about -90dB once time-aligned with a few touches of EQ...(if time-aligned, EQ should correct those phase differences also...though you might need some all pass filters also to get them all bang on)...
...btw I tested a few of these sub-sample delay plugins and the eventide was the most accurate I found...Airwindows has a free one but it's not accurate, "airwindows one is -12dB at 20k when at 0.5 samples...whereas the eventide is only -1.5dB. (at 48k sample rate)." ...if you just put it on every channel at 48k they'll have that same LPF...the airwindows one varies the LPF as you vary sub-sample time, so won't work for this...
Another cool free plugin that should help is Signalizer (jthorborg)...it's an oscilloscope plugin, very accurate, you can run it in Logic...small bit of a learning curve, check the presets and the youtube video...you can zoom right in and see both channels you're comparing and the sample points...with that and the eventide you can get things very close...(he hasn't updated it in years but it still works OK on apple silicon, works better on intel macs with a decent GPU though)...
Awesome work! That only note that I noticed was on the compressors you were testing Peak comp on the X32 vs RMS comp on the Avantis. Would that make a difference in your testing? Also are the comps on the Yamaha consoles RMS or Peak?
No it would not if he was appropriately mixing. It may affect the visual perception but if hes mixing by ear he could hear the compression.
amazing work... wonder how different then RPio sound... additionally, I wonder if there is a way to do thins same kind of testing to the “master buss” or the summing properties of each desk
It is nice to have some evidence before we talk about certain things. You just presented a lot of evidence!. Thank you for your time and passion!.
I found the exact same differences/points. I don't find the lack of "true 0 attack" thing to be a problem (i can always use plugins).
The thing is that with applications i have all these consoles can be made to sound the same. It's up to the features and user interface.
I'm used to Allen & Heath, X32 and Yamaha and would probably find working with Digico the hardest.
Is it just me or do the Digico consoles not seem to actually be in milliseconds? They even default very long. A whole deep dive video into their actual signal flow would be interesting. Is the “Mustard” after the SD if you run both. Are they parallel? If you use compressor on both dynamics, is it in series?
0msec Attack is really hard to do without buffering. A&H claims to have phase-aligned outputs all across the board (except when using FX slots), so it might be hard to be at zero-latency and zero-attack at the same time… while people complain that some Yamahas aren't phase-aligned but then again provide zero-attack compressors.
Behringer has some compressors/limiters in the FX rack that will do true zero-attack if you need to manage transients. I still haven't found a solution for my A&H SQ-5 yet, but also I don't have transient trouble too often anyways.
One thing is that consoles can sound rather different (in purpose) once you start using compressor models like in the A&H world, in the Behringer Wing, or in the rack of the Yamaha QL series.
Super great job! Can't help but to be curious what this could have looked like with the accuracy of 96/196kHz instead of 48kHz 😄
Thank you for all your effort. This is very, very helpful. God bless.
round two with digital in/out to hear even less difference ! and don't you have a pm3500 or h3000 left in a corner instead of the mackie ?
thank you very much for your work !
Seems like the high frequency content of each desk is what makes it different
Minor quibble. Q is not "squelch". It's a standard named variable in filter math. Great work all around. I really enjoyed this video.
True, good point. I’m guessing that like me, most sound engineers have come to use those interchangeably because they sort of point to the same effect
@@devinlsheets_alphasound Interesting. I've never heard or seen it referred to that way. Where are you located? Squelch in rf is more like a gate. Mute this audio if I don't have enough carrier signal.
Q factor
@@NotTooLoud That is correct. Squelch has to do with noise suppression in RF circuitry whereas 'Q' is more closely related to bandwidth of filters. In my 40 years of working with audio, both as a mixer and technician, I have never heard these terms used interchangably. It's a minor quibble for sure. Otherwise, the content is great. I certainly appreciate the time it took to pull this off.
Thanks for this!
Very Interesting, I've heard people say that different consoles sum/mix the channels differently too, I don't think I've seen sometime test this. Maybe run multitrack though each console and see what the master output sounds like on each for various styles of music?
yeah people buy analog summing boxes for this reason, worth testing
dan worral has something on summing, not exactly what youre looking for but it's related and dan worral rules th-cam.com/video/wVp4syrFkE0/w-d-xo.html
You could but it really doesn't work like that. He covers the factors here, any changes would be additional, not multiplied. Same thing has been done with DAW's for a long time as well. In the world of DSP it doesn't matter.
@@ts4gv People also buy $10k IEC cables. And I do believe those can make it sound better *to them*, because everything affects your perception of sound.
next time would be super interesting to see you inlcude a waves server rig so compare to the stock console processing
Can you please try adding a Nyquist filtered version reference track to the red pulse test, since that part of change looks significant on time domain but is almost inaudible for human? I guess the untreated uncompressed console output well match the Nyquist filtered version of reference very much.
Great info ..and no muzak spoiling it as well A++
I was really tempted to put in background music just for you
Really interesting video. From your snat tests on the compressors, it's clear that only the MTX5, MRX7 and TF are truely going to 0ms, evidenced by how they suppress the initial click. All the others, when set at 0ms, are letting the initial click through, except for the DM0 which is partially clamping the click. Although accurate to the user setting of 0ms, this behaviour on the MTX5, MRX7 and TF will sound a little more distorted - and I think this was apparent in your 1/2 cycle tests. Also, the Quantum seems to have a very wierd compressor algorithm. BTW please don't use "squelch" for Q or bandwidth. :-) Squelch is related to suppression or gating - like when a RF receiver suppresses the audio output when the received RF level is low.
Very interesting test, thank you!
Sorry but i really liked the color of the compression from the digico on that drum loop :p .... even though, i get it, that wasn't the purpose !
I would prefer that the standard compressor not have that kind of alternation to the tone. It should be clean. There could be a separate mode or setting that adds tone, and then distortion is welcome! But to not have a clean option?
I was right there with you until the kick/snare/hat loop playback part. Why did it sound like the levels and perhaps a high pass filter were on some of the consoles? Was those the ACTUAL differences between them? I must have missed something. Also, labels on the channels you were unmuting would have been cool, to keep track as well as your narrative. You tested these things exactly the way I would have. But I don’t use a Logic interface. I simply use Studio Six Audio Tools, with a pink noise reference, reproduced as needed, Y-cables if needed, to check drivers and consistency across my arsenal. I have used this method to realize why my wedges often sound different, where I discovered a very real issue for the first time in almost 30 years. I removed the diaphragm from one wedge that I had graded a “C”, for fidelity. I had lots of “A” grades, one “A+”, which I cannot explain, no “B”s, and two “C”s, one from the 12” range, and the other being my newly realized discovery. Which might sound dumb and obvious but previously had not been. There was crud in the voice coil groove of the HF driver. Like dark, sandy, gritty, semi-sticky crud. Masking tape revealed nothing, but there was crud on the voice coil itself, but not uniformly around. Hmmm. I’m in Charleston, SC, quite a humid place, with average relative humidity being 78%. Generally, after the summer onslaught of gigs, by around October or November, I used to find some failures, which turned out to be corrosion on the diaphragm, which apparently caused an imbalance enough to allow the coil to hit the magnet somewhat, and eventually would sever the copper, rendering no continuity, yet was imperceptible to the naked eye. After three years in a row of observing that, I started keeping fresh Damp Rid hanging with cabs in a somewhat sealed environment. And it has worked wonders. Wish I had done that with my mic boxes during Covid, where I lost three SM58s, and one 56, presumably to corroded diaphragms. Ugh. But back to the crud. Blue shop towels didn’t even grab the the stuff. BUT, a medium to heavy guitar pick wrapped in an alcohol soaked blue shop towel, grabbed it well. I imagine that’s in more gaps than we would think. I take care of my gear, but they stuff gets used, and if my gear can’t handle the real world, well then it won’t be my gear for long. JBL and EV make the best drivers, which is why Clair Bros. and EAW used JBL drivers in all of their development mules and even production cabs for many years. But what I had thought were damaged crossover networks or torn diaphragms were neither. Just crap in the gap, and taking about 1/2 hour per driver made them sound like “A+”s again! An “A” grade was awarded by my ear, then a loaded up 58 through them as monitor wedges, pushed hard, and by my ear, then by RTA monitoring pink through the wedges, where I took the best sounding box to my ear, which meant thick and rich and chocolaty, hifi, and serious gain before feedback. That “curve” became my reference to grade the others. Age differences in the 12”, their surrounds, and even the weight of the paper, which did affect the phase even above the crossover point, appeared to be the only causes of the different grades. One db shy at 3.15khz, or two at 10khz, or three at 315hz isn’t going to ruin my day, but your comparison was so close to my method, I had to let you know how I used it. As far as consoles, I’ll tell you what I’ve observed. I had not realized how much of an audio snob I had become. I had a Verona and an XL3 monitor board, with the Midas XL3 being likely the greatest console ever made for live. The pre’s simply would not break up. I had renowned, world touring musicians using In-Ears, quite the telling reference themselves, walk over and ask me what I was doing to give them such a great mix. Every time I just pointed to the mighty Midas. Then the cheaper, smaller digital consoles started coming out. I had an O1V/96 back in 2003 or 2004, and when a mic stand tipped and broke the screen, that was my excuse to throw it in the trash. Those pres were junk. All Yamaha consoles, from M7 back, make inexcusable noise, regardless of load. If you can get over that, they rock, and they are famously versatile, but so are Allen&Heaths. Great EQs, so-so pres. My Venice was my baby. Then I screwed up and bought the Presonus Studio Live thing. Whatever. The day the X32 came out, I flipped the Presonus to a church who simply wanted recallable scenes, saved them a lot of money, and still paid less for the X32. Win, win, win. If you are making an X32 pre break up, you need to grasp what a hard red light means. I’m still not sure if I’ve ever heard one breakup, ever. But ironically, I did hear an M32 get rough on very loud vocals, and I even checked to Shure RF receiver, and it was not even touching clip, and the vocal channel pres were, but not slamming, meaning not hitting red well before the amplitude had reached peak. The distortion was slight, but there. I still don’t get it. I understand from people “in the know”, that the M versus X version uses “higher quality” gain stages by the third or fourth stage in the M32 preamps. So up till, let’s say -20, or -15, I guess, they are identical preamps. Either way, I can’t tell and I’ve had them side by side. It was very surprising to hear an M32 “get rough” though. I’m an old school Soundcraft fan, but those Expression desks are fine, if you don’t push the pres and you don’t allow too much gain reduction on the channel comps. Those truly distort. I understand that compression does technically induce distortion, but it’s usually resolved before leaving the device, like a dbc166XL. I installed a DiGiCo S21 a while back, and that honestly was a pleasure to use and touch, with some details which were amazing, but it was TOO pristine, too clinical. No balls to the pres. But right when the X32 came out, I also bought the Mackie DL160whatever. As a toy/under the arm utility in a pinch. Used it on top of my Verona for opening bands where I didn’t want the Midas touched. Always had to mess with the Mackie. Always. Always Eq’ing, but never satisfied. It took a while to sink in. One night I used the DL Mackie just to see what was up, and had the entire Verona to use also. Soundcheck sucked, so I simply plugged into the Verona, a true A-B test, and my jaw dropped. Like butter that Midas was. I only needed to do the usual high pass on Fox and OHs, HHs, and that was it. An epiphany of sorts. And what a spoiled snob I realized I had become. The same guy who criticized sound snobs. Whoops. Anyway, your comparo was very, very cool. There are differences, but it’s more in dynamics than static settings. As you noted. Cheap Yamaha “blue” mixers, can’t be pushed anywhere NEAR clip before you hear it. And I love Yamaha. Great pianos, motorcycles, outboards, home stereo receivers, lawnmowers, and even legendary FX units. But ALL Yamaha audio stuff is noisy. I can’t reconcile why. The newer CL and TF stuff is fine, but not intuitive in use for me. Fine, but maddening. And I know the M7CL damned well. Why is the new stuff such a pain? If it isn’t intuitive to me by the fifth or tenth use, it’s not intrusive at all, in my book. DiGiCo, is. No problem. Avid? Give me a break. ProTools for the road. To me, those desks are for the college kid with a slush fund who bought his Apple G5 and ProTools and likened himself to an audio engineer who could take his million takes, over manipulated, hyper quantized, few acoustic variables, very little mic bleed, mixes, and do them live. The results are so sad. It sad to watch a guy struggle to try to do that controlled environment studio stuff live and just not get it, then blame everyone and everything but himself for the problems. Hey man, room modes exist, feedback exists, mic bleed exists, and one take is the real world. Studios are the fantasy world, and for good reason. We all want to hear the magical results of those fantasy worlds. Siamese Dream, anyone? A sound to revere as much as any other. Incredible. Live? Only with track accompaniment. It takes a truly exceptional band, engineer, and set of equipment to make live sound like “the album “. Check out Gojira. They are one of those. Live mixes rivaling studio mixes, EVERY TIME.
One take. That’s the real world.
Perfekt vid for earls morning drive with studio headphones. ☀️
Driving with studio headphones on. Bold
Thank you for this.
There are differences. Nice to see them quantified.
The video was great.
Can you please make a video about the older Yamaha consoles. Like the DM series and 02R96, which was heard a lot about its preamps.
Send them to me!
Same Preamps as the other ones from that time... There's so much "heard" from people with little understanding, don't waste your time.
Not same preamp.. look the service manual and schematics. I’ve got two DM2000, one 02R96 and a 01V96..
Also converter are different
What do you think about the presonus studiolive 32sx
Not really what I expected either. Would love to see a follow up with multi-tracks to hear the summing, then with basic EQ, then with basic/simple EQ and Comp settings.
Working at University Of Texas at Austin I first had a 01v96 which I literally saved from surplus. It was used in a lecture hall and I had to twist my
supervisor's arm to let me claim it and use it on gigs. He had used analog gear since the 50's and was very suspicious of this new-fanged digital stuff. He never
did get the hang of digital but the LBJ Library eventually installed an M7 (I loved it, and eventually, so did he...all those faders!) Our group eventually
ended up buying two L9s and several Allen and Heaths and a mess of old analogue boards ended up banished to surplus.
Great work 👏👏👏👏👏👏👏
Great Test. Would've been nice to see how the Behringer Wing would've fared in this shootout.
What can you expect? It would use the exact same preamp option as the XM32-Family.
The sound-processing on the other hand is much more versatile and on par with "high-end"-consoles.
@@hafibeat834Doesn’t the Wing utilize the Midas pres for the onboards?
@@TheDude1764 Yes, but only 8 of them.
@@hafibeat834 Ah…that’s right. But depending on what stage box you select, you can have the choice of Midas pres or the Behringer pres…as is the same with other X32-family consoles I suppose.
@@TheDude1764 yes, I wrote that before. But the difference is too small to hear in a live situation.
The "real" Midas Preamps are very very good tho.
The drums and vocals sound so realistic
IMHO, the reason why most.all digitally controlled preamps sound the same is because they're using the Texaa Instrument's PGA2500 chip (a transparent, quiet, clean chip). It will be interesting if you can find a console/preamp that's using THAT Corp's digitally controlled preamp chip, which was introduced years much later. Thanks for this demo, fascinating!
Were all these digital components synced to a master word clock? I might have missed that point.
Would love to see a test with the new Midas HD96
Great job!
Thank you John for the O1V96i and the Quantum 225!
There are distinct differences in the sound of these consoles. The older Yamaha stuff sounds less dynamic and midrange focused.
This was really useful for me! I'm running an M32R and looking to upgrade to something with newer tech... My M32R sounds vastly better than the Mackie Onyx it replaced with much higher clarity and better "richness" to the sound. With the null tests, anything else nulled to the Mackie had some low and high bleed, which indicates to me that the Mackie has a slightly narrower bandwidth than the other consoles.
All that said, the Yamaha consoles, as you point out, seem to be the best behaved, particularly with compression. I was really disappointed that the really expensive DiGiCo would have that kind of distortion in its compressor. (Fortunately, since that's just a software algorithm, a firmware update could fix that.) What encourages me is that, fundamentally, these consoles sound pretty darned similar, and so I wouldn't lose anything, sound quality-wise, by going to, say, Yamaha. In fact, since its compression is better behaved than the X32 compression, I might actually get better sounding shows. The DM3 is interesting for such a compact console, and the DM7 is drool-worthy... I'm hoping for a DM-5! That'd be pretty sick.
I don't think you will get better sounding shows using something other than an x32 so I wouldn't worry. There isn't much you can actually "upgrade" about u
@@marcusschulte1342 my M32R sounds brilliant! Zero complaints with the sound. I was actually concerned that other boards might be a downgrade from this sound quality, and I'm looking to get into something larger. But I think I don't have to worry about that (despite the Yamaha reps telling me I'd have to "work a bit harder" to get that kind of sound on a Yamaha board below the Rivage/RPiO hybrid preamps).
@@soundman1402what you are looking for is the Behringer Wing then. Also if your rep is talking to you about the "preamp quality" in the boards then it's probably time to find a new rep.
subscribed, great video
Good Video - but what exactly is the surprise?? 🤔
46:15 conclusion
Like to see digital vs analoge in case of compression, limiting and dynamic mic comparement. And mixing bass and heights at live without trashing ppl ears, plus correct phase alignment.
How would Tio1608 and Rio1608-D2 compare? Especially in terms of noise floor?
Ok, but we have to keep in mind that smal variations will sum up with many channels and many process working in each of them. So...what is not expected small difference at the end of 60 inputs can sounds spectacular different than we expected :)
These days electronic components are so good that difference betwen great and good one is probably 5% - this cheapest one manufactures keeps less attention to elements selection than this high price products.
That artifact at 32:23 almost sounds like the classic Yamaha FM synthesis
Question: Are you clocking all these from their respective internal oscillators? Or are you providing an external time reference?
Tried both ways. Doesn’t seem to make a difference.
@@devinlsheets_alphasound Ok interesting.
I have been told by some engineers that clocking the Yamaha CL/QL from external word clock will have a noticeable improvement on the mix sound. I haven't tried this personally, but that is why I was asking.
I believe a lot of the „sound“ of the consoles happen when summing multiple signals.
I have not yet come up with a good way of testing this…
Mmmmm. You could have a mix on a daw, then route it multitrack to the board with all faders at unity. Then compare the summing of all.
In 2012 a group of Dutch sound engineers got together and compared several digital consoles, from X32 up to some very high priced ones.
Conclusion:
Differences between consoles were pretty difficult to hear in a otherwise completely quiet room with just playback, let alone in a (small) crowd of people.
Agreed. The only exception where consoles matter, in my opinion, would be something like a massive festival or big-name concert where, indeed, a big sound system could reveal the flaws of cheaper equipment. For small bars, churches or live venues, an mid-range console is more than enough.
@@Max16032 "a big sound system could reveal the flaws of cheaper equipment"
This is obviously not the truth for many reasosn. This test here reveals it all -> nothing to write home about.
I'd love to hear a groovy drum loop played under heavy compression as well!
This is an awesome video! Shame that you spent so much on all of them and had to skimp on the camera quality 😂. Thanks for the video though, very interesting and clearly very well thought out and put together
why not just to feed the noise directly without any microphone at all to bypass everything besides the preamp itself?
Thanks for an excellent comparison of so many consoles! Very thorough! I noticed that during the compressor comparisons, the Avantis was in RMS detection mode-I'd be curious to see if it still has the same transient response when set to Peak detection
Good stuff!
I went from several years day to day on an M7CL to a CL5. Same job same same venue. Immediately noticed the CL5 sounding significantly better. Other people commented as well.
Yeah we hear stories like this a lot. It’s totally possible that there is a difference in sound in these situations. But it isn’t clear where it comes from if so. It seems strange that it wouldn’t show up in these kinds of tests.
@@devinlsheets_alphasoundWouldn’t it be reasonable that the preamps are going to ‘wear’ over time and the change to a new desk is going to restore some of the lost brilliance?
@@TheDude1764 not sure that the specific electronics in a pre-amp are particularly influenced by age, at least not in the lengths we're dealing with. Maybe after 100 years lol, but ten or twenty years doesn't seem long enough to make the M7 or LS9 sound all that much different than the CL and QL. Or, at least, whatever difference may be caused by time, it's so little that it barely shifts the null test away from zero.
22:05 It's honestly ridiculous that the Avantis is arbitrarily limiting the release time to be no faster than 50ms
older A&H consoles don't even go below 100ms and I very rarely have a compressor with a slower setting than that..
it's like the X32's low-cut not going any higher than 400Hz... limiting my workflow and making my job harder for no reason at all
Also, you've set the Avantis compressor to 'Manual RMS'. It would need to be 'Manual Peak' to do the same thing as the other consoles.
27:55 What you're seeing is not compression, but it's a change in badwidth; the console isn't DC coupled and also doesn't extend into infinity, which is why the return signal looks different on the scope
29:25 This confirms my suspicion that the X32 doesn't get super quick with the attack time; this has annoyed me several times in the past
29:39 No console goes to literal 0 attack, unless they have look-ahead, which is only possible by delaying the signal ("seeing into the future").
Most of them probably show any attack time below 1ms as '0', but that still means one console might be 0.5ms, while another might be 0.1ms while showing a zero on screen.
37:57 A compressor in a non-linear effect and non-linearities are simply distortion; distortion is inherent to a compressor. A compressor with literal 0 attack and 0 release time is hard clipping (like digital clipping). The faster the compressor works, the more audible distortion it generates. The differences we're hearing between the consoles might simply be difference in attack times and since they're all rounded to 0, it looks like they're all the same, but probably vary a lot.
On that note: compressor timings are also not standardized. A 50ms release time might sound different on one compressor than on another one. Does it drop 3dB of gain reduction per release time? 10dB? All of it? Does it drop linearly? Logarithmically? These are all things you can tweak in some studio plugins, but are mostly a black box when it comes to hardware units. They sound different because they function differently. Not because one it better or one it worse, but because the settings are different, even if they 'look' the same on screen.
42:30 I wonder if this is due to an incorrect decramping implementation. If so, it'll probably be fixed in a future firmware update.
43:39 Post ringing is an atrefact of linear phase EQ/processing. The channel EQs on all your consoles is minimum phase EQ, which does not cause ringing. The difference we're seeing on the scope is (again) just a function of frequency response. Frequency response and impulse response are related (or the fast fourier transform wouldn't be possible!) so the waveform on your scope will look different if you change the frequency response of the DUT.
All in all very cool and informative video and while the null test is a cool way to illustrate differences and give the viewer a sound example, it's not the most scientific of tests. You could get REW running on a PC and do all kinds of tests. Frequency response, THD+N, Dynamic Range, IMD, SINAD vs Level; there are so many tests you can do! I thoroughly tested my X32 this way and published the results on audiosciencereview, but I'm not gonna link the thread again, or my comment will probably just get deleted by youtube again, like the first time I typed all this. I hope I didn't miss anything, but I think I probably did. Anyways, thanks for the video and keep up the good work!
I should flag this comment just to be today’s biggest troll. HAHA. Just kidding, these are excellent comments and I’ll try to find the time to respond sometime
Ok so the post ringing. The EQ’s on the consoles are not linear phase EQ’s. But they ring… although, they only ring after the sound impulse. Linear phase EQ’s seems to ring both before and after the impulse. No?
@@devinlsheets_alphasound Minimum phase/IIR EQs don't ring. The longer wave form you see is the result of the change in frequency (and thus phase) response. The impulse response (/step response) graph is probably one of the most misinterpreted things in audio. Looking at the impulse/step response is basically a worse way of looking at frequency response.
Can a 0ms attack compressor even be distortion free? The closer to 0ms the more you are converting the transient into a square wave.
With digital lookahead, there can be a zero millisecond attack that is distortion free. The problem is how to get as little lookahead latency as possible for live applications. Less lookahead means pulling more digital tricks to avoid distortions. Yamaha decided to try a truly zero attack on some units but there is some amount of distortion, especially at lower frequencies. They probably found a balance they thought was appropriate for those units.
You are right, just use a slower time.
Round 2, please!!! LS, QL, DM3, X32, Avantis, Digico sound test with actual instruments and/or vocals plugged into the consoles themselves through a PA and DAW. Let’s test if the higher fidelity console’s really do stand out or not.
A test of their recording capability would also be awesome. Example: 48k vs 96k does it really matter? Are the X32 preamps warmer? Are the Yamaha’s colder and fully transparent in comparison? Can the difference even be heard by the human ear? I’m willing to bet there is not, and it’ll just be personal preference of what we as individuals like to hear and work with.
Thank you for this test. Confirmed and debunked a lot of information floating around the internet about all of these consoles. I was also surprised at the amount of similarity vs differences between them.
From a budget standpoint of say 1k-3k, I see no reason why a used LS9 or newer DM3 would ever be passed up. Build quality, portability, workflow, etc. X32/M32 is also in this range but using an X32 Rack in the past, I was not impressed and the effects were also lackluster, imo. I’d imagine effects would be better in the Yamaha consoles. X32/M32 probably stands out the most in terms of routing and flexibility. The Avantis and Digico are just too large of a format for me and price point is just out of my range. Same for QL, even though the QL1 is probably super portable and can do anything/everything these other boards can.
Just think about how many churches have been upsold on consoles by people claiming that the more expensive ones, or some particular brand at any price level, has a better inherent "sound" that they need if they're a serious client. It's total crap. There ARE reasons to buy more expensive consoles, or perhaps those of a certain brand, but significant tonal differences in preamps or basic processing doesn't seem to be one of them, at least among these digital pieces. The funnier part is: how many clients have convinced themselves that they DO hear big differences in these areas because they either really wanted to, or they paid so much money that they now have to?
@@devinlsheets_alphasound 100% agree and your test, for me, confirmed that exact same notion. I’ve learned most of the keyboard warriors out in the forums/blogs are mostly talking nonsense. I ignore most of it now and check into things that do come across as great advice and run with it. Rare though.
I’ve also seen some crazy rigs out there and even engineers with the latest and greatest but at a significant $$$ and just wonder, why? I’ve also seen engineers with nothing more than a M32 or PRO1, or older Yamaha’s like LS9, 01V96 with just a stage box. That’s it. Phenomenal mixes. Guys with a 15k rig in a crate couldn’t touch it. A better tool will always help but really it’s down to the user and what they can do with it.
I know the big thing right now with most churches and where they get most caught up is with live streaming and/or recording. So, the tool does matters here and some will make things easier than others but you don’t have to break the bank for it and “major” tonal differences are bogus.
"48k vs 96k does it really matter? Are the X32 preamps warmer? Are the Yamaha’s colder and fully transparent in comparison? Can the difference even be heard by the human ear?"
No, No , No, No and No. A waste of time.
@@hafibeat834 It seems to me that you still haven't tried cooking on the console screen. The x32 cooks the egg faster.
Waves LV1 was missing, I would have liked to have seen the comparison with LV1!
Send me one!
Welp that settles it! I'm keeping my Mackie 1202.
Kidding, sort-of. I enjoyed my Soundcraft Si Impact for several years but am retired now.
Thanks a lot
Since you were using a DAW, why didn't you just nudge the regions of the recorded consoles outputs to time align them? The number of plugins you have assigned and engaged will affect the latency for each channel. Latency compensation inside of DAWs It's not exact.
He did that. That's exactly what all those bits where he's adding or subtracting 1 sample were.
great video! i noticed at 1.51 the Avantis tone is sharp compared to the others by a good half a semitone or more
Or just brightness
He said in the beginning that Avantis can't do 1kHz exactly
The Avantis is just too special for 1kHz
@@devinlsheets_alphasound For people with special needs?
Thanks
From the thumbnail I thought it would be beach chairs test
Great, but few comments.
Differences on compression could be due different latency on effect process in each console.
You are not comparing apples and apples in same time.
Came here to say this. Also, would have loved to just hear some full-range music at the end, but maybe TH-cam would have flagged it. Thanks for taking the time to do this.
Dang I wish you had included a Presonus SL console. and no one is going to do all this again any day soon, Man I would have e sent you one to include.
Thanks for all your hard work not only in the audio tech arena but the great narration as well as video editing and publishing, the entire package was really well done. and some interesting info learned here.
Just stumbled upon your channel here and will be checking out other works.
The Yamaha is such a ripoff aesthetically of the PreSonus StudioLives.
@@cliffprowse3341Yamaha has been making digital mixers for much longer than PreSonus.
The Avantis should have had the Manual Peak compressor recalled not the RMS. So I would say that the Avantis tests need to be redone.
What do you suspect would be different?
@@devinlsheets_alphasound I would think the compressor would react faster on peak mode than RMS mode, am I wrong?
Not sure, would have to test it!
@@devinlsheets_alphasound Well I just checked the pro tools 'pro compressor' (its the only one I have available with switchable peak/rms detection) with 100microsecond attack (fastest) and 50ms release on a short 1kHz sine, the difference is really big. You might want to check the Avantis again, if possible... Besides, I own a dLive and I find myself using the RMS compressor for almost everything that I want to compress in a 'normal' way. When going for 6-8dB or more of compression on a vocal with fast attack and release I find the Peak compressor to sound a bit thin, choked. The RMS compressor sounds better imo. I never A/Bd them, I should though.
@@BobBriessinck yes interesting! I’m getting the feeling that a second round of testing may be needed. Who wants to fund that project??? lol
Is it okay for me to understand this from this video? 1) No matter what console you use, there is little difference in sound quality. In other words, there is no difference between x32 and digico quantum225. 2) There is no difference between the various processors of the console. // Should I understand it like this? If I have a question, even though there is no difference in processor process, doesn't the tone difference of the preamp sometimes feel? For example, when comparing x32 or digico quantum225.
Any reason why the Tascam Model 24 is not there?
You didn’t send us one
What a pitty you didnt run a recording of a track (wewill rock you, something with low end) through the chanelstrip and rerecorded it 10 times to see if anything changes…
thanks!
Jesus this video is EXPENSIVE
💚💚💚💚💚💚It, fantastic, Thank you,
Now weed to compare Walmart lamspeaker cable with a known AWG VS Audiophile $$$ voodoo cables
Cool idea to compare consoles like that, but unfortunately you're bottlenecking the signals by running into rio, no wonder higher fidelity consoles can't stand out, their signals are being chopped. A much better way would be to have a mic'ed instrument running in to a good quality splitter and then to preamps.
Also, higher end consoles have more dsp meaning they are able to do more complex calculations which will play out when summing channels. That, combined with better dac leads to a drastic change in headroom, like a whooping leap from +18 dBu on x32 to +24 dBu on SD rack.
Lastly, would be amazing to have waterfall frequency responses, something that is seemingly being completely ignored by manufacturers because it wouldn't allow them to brag about "look how linear our frequency response is".
Thank you for this test, it's quite amazing that you took all that time to do it for all of us!
sounds like we need a round two, eh?
@@devinlsheets_alphasound If you used the Avantis as the destination, you'd have 96k to do the time alignment.
@@NotTooLoud nice. Also, Logic will do 192 but I don’t have an interface that will do that, and, I need to also have an analyzer that will do 192. And even if Logic and the analyzer and the I/O do 192, I need a digital router that will do that. Dante Via won’t do 192 I believe, and DVS is too limited in terms of routing to do what I really want.
Why is Avantis different in tone? at 2 minutes...
Why was the avant is pitched up?
Because it’s super special